Workflow Explanation

This repository contains a set of Jupyter notebooks to handle and preprocess data efficiently. Below is an explanation of the workflow and purpose of the included notebooks:

1. Downloading the Data

• Notebook: DownloadData.ipynb
• Purpose: This notebook is used to download all the necessary data required for analysis. It ensures that all datasets are collected and organized in the appropriate directories for further processing.

2. Reading and Processing the Data

• Notebook: ReadingData.ipynb
• Purpose: This notebook:
  • Reads all the downloaded data files efficiently.
  • Processes the data to create training and testing datasets for antibody titer analysis.
  • Saves the following CSV files:
    • abtiter_data_X_train.csv: Features for training.
    • abtiter_data_X_test.csv: Features for testing.
    • abtiter_data_y_train.csv: Labels for training (not available for the test set).

3. Task IGg_PT Folder

• Location: CMI-PB_SparseAutoencoder_RandomForests\Tasks\Task_IGg_PT
• Contents: This folder contains:
  • Data: Includes the processed datasets for the IGg prediction task.
  • Scripts: Contains code for model training and prediction.
    • Key Script: SparseAutoencoderRegression_v1.ipynb is the main script used for running all predictions.
  • Results: Stores the outputs and results from the prediction models.

Key Notes

• The training labels (y_train) are only provided for the training set. -Submitted predictions only for antibody titer task. But preprocessed the data for monocyte task as well